HIV protease inhibitors are an important new class of drugs which have made a significant impact on the health care of AIDS patients since the first one, saquinavir, was introduced to the marketplace in 1995. Examples of other protease inhibitors include amprenavir, indinavir, nelfinavir, lopinavir, ritonavir, and atazanavir. They are especially effective in combination with other anti-HIV drugs such as reverse transcriptase inhibitors or with other HIV protease inhibitors. In spite of remarkable success with these new therapeutic regimens, there are strong indications that results would be much improved if therapeutic drug testing methods were available for monitoring the concentrations of protease inhibitors. Not all patients respond optimally to protease inhibitor combination therapies. Even those who do respond can subsequently develop drug resistance due to the notoriously high rate of mutation of the HIV virus. However, it has been shown that there is a clear relationship between plasma levels of the protease inhibitors and therapeutic efficacy based upon decreased viral load and increased CD4 cell count. One problem lies in the fact that the drugs are metabolized extensively and are subject to complex drug-drug interactions. The results are extremely complex pharmacokinetics and a strong element of unpredictability between dosage and resultant drug levels at any particular time for any particular patient. With therapeutic drug monitoring, drug dosages could be individualized to the patient, and the chances of keeping the virus in check would be much higher. But routine therapeutic drug monitoring of protease inhibitors would require the availability of simple automated tests adaptable to high throughput clinical analyzers. Currently most reports on therapeutic drug monitoring of protease inhibitors have used HPLC methods which are slow, labor-intensive, and expensive. Recently there was a report of a radioimmunoassay (RIA) method for saquinavir (Wiltshire et al., Analytical Biochemistry 281, 105–114, 2000). However, such a method would not be adaptable to high-throughput therapeutic drug monitoring and, like all RIA methods, suffers from the disadvantages of having regulatory, safety and waste disposal issues related to the radioactive isotope label used in the assay. The most desirable assay formats for therapeutic drug monitoring are non-isotopic immunoassays, and such methods have heretofore been unknown for monitoring HIV protease inhibitors.
As indicated above, HPLC has been the method of choice for monitoring HIV protease inhibitors. Two recent reports in the literature describe HPLC assays for the simultaneous determination of several protease inhibitors in human plasma, Poirier et al., Therapeutic Drug Monitoring 22, 465–473, 2000 and Remmel et al., Clinical Chemistry 46, 73–81, 2000.
Chemical and biological assays generally involve contacting the analyte of interest with a pre-determined amount of one or more assay reagents, measuring one or more properties of a resulting product (the detection product), and correlating the measured value with the amount of analyte present in the original sample, typically by using a relationship determined from standard or calibration samples containing known amounts of analyte of interest in the range expected for the sample to be tested. Typically, the detection product incorporates one or more detectable labels which are provided by one or more assay reagents. Examples of commonly used labels include functionalized microparticles, radioactive isotope labels such as 125I and 32P, enzymes such as peroxidase and beta-galactosidase and enzyme substrate labels, fluorescent labels such as fluoresceins and rhodamines, electron-spin resonance labels such as nitroxide free radicals, immunoreactive labels such as antibodies and antigens, labels which are one member of a binding pair such as biotin-avidin and biotin-streptavidin, and electrochemiluminescent labels such as those containing a ruthenium bipyridyl moiety. Sandwich assays typically involve forming a complex in which the analyte of interest is sandwiched between one assay reagent which is ultimately used for separation, e.g., antibody, antigen, or one member of a binding pair, and a second assay reagent which provides a detectable label. Competition assays typically involve a system in which both the analyte of interest and an analog of the analyte compete for a binding site on another reagent, e.g., an antibody, wherein one of the analyte, analog or binding reagent possesses a detectable label.
Copending U.S. patent application Ser. No. 09/712,525 filed Nov. 14, 2000 having the same assignee as the present application and published as EP 1 207 394 on May 22, 2002, describes a non-isotopic immunoassay for an HIV protease inhibitor comprising incubating a sample containing the inhibitor with a receptor specific for the inhibitor or for a metabolite of said inhibitor and further with a conjugate comprising an analog of the inhibitor and a non-isotopic signal generating moiety. Signal generated as a result of binding of the inhibitor by the receptor is measured and correlated with the presence or amount of protease inhibitor in the original sample. The protease inhibitor conjugates of the present invention are especially useful in such an assay.
Among other problems, there is a need for an antibody specific for atazanavir having essentially no cross-reactivity with the HIV protease inhibitors saquinavir, amprenavir, indinavir, nelfinavir, lopinavir, and ritonavir or with the atazanavir metabolite 4-pyridin-2-yl-benzoic acid.